Read the mean, median and distribution in StoryFlop
Calculated examples to understand what votes describe and which decisions remain with the team.
Explore a round's results
Compare prepared examples. Calculations follow StoryFlop's rules; they do not decide for the team.
- Numeric mean
- 7
- Numeric median
- 5
- All votes match
- No
- 31
- 51
- 131
Special cards appear in the distribution but not the mean or median. Zero does count. Matching cards are not a documented agreement.
1. Start with the cards, not the average
The distribution shows how many people chose each card. Before a summary number, check whether votes cluster, an extreme stands out or questions appear. The same average can hide very different conversations.
These fictional examples reproduce StoryFlop's calculation rules. Numbers help read results, not convert points to hours or compare individual or team performance.
2. Mean and median answer different questions
For 3, 5 and 13, the mean is 7: numeric votes added and divided by their count. The median is 5: the middle ordered value. The mean need not be a deck card.
Neither proves agreement. Ask what the 3 included and what the 13 saw. If they understood different work, clarifying scope matters more than choosing between 5 and 7.
3. An outlier deserves a question
For 5, 5, 5 and 21, the mean is 9 and median 5. The median describes the centre without removing information in the 21. That voter may know an overlooked dependency.
Ask about work and risk rather than requiring a defence of a number. If information changes shared scope, vote again; otherwise record an investigation or split.
4. Zero counts; questions are not numbers
For 0, 5 and ?, the numeric votes are 0 and 5, so mean and median are both 2.5. Zero participates. The ? stays in the distribution and asks for clarification; it is not zero.
With ? and ☕ there are no numeric votes, so no mean or median is available. Even identical special cards express a question or break, not an agreed work estimate.
5. Separate the summary from the final decision
Matching cards mean identical votes. A useful agreement also confirms scope, quality and assumptions. Record why a final estimate was chosen and review it when the story changes.
Use history to recall which information caused another round. Do not rank people by distance from the mean. Planning Poker should help understand work rather than reward conformity.
Put it into practice
Sources and further reading
The examples and StoryFlop instructions were written for this section. These sources provide context for the method and refinement: